EUVIMEDEUVIMED entdecken
Crossref

A user classification approach based on anthropicity for counteracting AI bots on interactive online platforms

T.I. Korobeinikova, V.V. Suslov

Radiotekhnika · 25.07.2026 · (225) · 76-83

DOI: 10.30837/rt.2026.2.225.06 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Radiotekhnika, online erschienen am 25.07.2026. Verfasst von T.I. Korobeinikova und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper addresses the problem of counteracting AI bots on interactive online platforms, which increasingly imitate human behavior, rendering traditional passive detection methods (CAPTCHA, behavioral analysis, reputation systems) ineffective. An approach to user classification based on anthropicity is proposed, which relies on a paradigm shift from passive detection to active verification. The approach involves a one-time verification of a subject’s human origin through the execution of adaptive tasks at three levels of complexity: EASY (object recognition), MEDIUM (emotional intelligence), and HARD (embodied intelligence). For quantitative assessment of anthropicity, an integral indicator S is introduced, which takes into account answer correctness (C), reaction time (R), and answer change (A). An experimental study involving real users and three AI models (ChatGPT, Gemini, Claude) confirmed the effectiveness of the proposed approach: the AI bot detection rate reached 93%, which is 31–48% higher than that of conventional methods, while the false positive rate was only 3% (compared to 8–15% for existing approaches). The proposed approach can be implemented in interactive online platforms to ensure protection against AI bots.

Zitieren

APA

Korobeinikova, T., & Suslov, V. (2026). A user classification approach based on anthropicity for counteracting AI bots on interactive online platforms. Radiotekhnika, 76-83. https://doi.org/10.30837/rt.2026.2.225.06

Vancouver

Korobeinikova T, Suslov V. A user classification approach based on anthropicity for counteracting AI bots on interactive online platforms. Radiotekhnika. 2026:76-83. doi:10.30837/rt.2026.2.225.06

BibTeX
@article{Korobeinikova2026,
  author = {Korobeinikova, T.I. and Suslov, V.V.},
  title = {A user classification approach based on anthropicity for counteracting AI bots on interactive online platforms},
  journal = {Radiotekhnika},
  year = {2026},
  number = {225},
  pages = {76--83},
  doi = {10.30837/rt.2026.2.225.06}
}
Kostenlos starten →